I asked ChatGPT the following question:
"In the NHL, they talk about underlying numbers to support or discredit a teams record. For example, the Ottawa Senators are said to have great underlying numbers but their goaltending is causing them to lose games. Can you get me the Toronto Maple Leafs underlying numbers and put them into context for where they are in the standings?"
It gave me this, which I think is saying their record should be better than it is:
Here’s a snapshot of the
Toronto Maple Leafs’ 2025-26 season performance and underlying numbers (as of mid-January 2026), and what those numbers
suggest about their actual play vs their results:
- Record: ~24-17-8 (56 points) in 49 games.
- Division: 6th in the Atlantic Division.
- Points pace: Competitive but not dominant — outside a top wild-card spot and trailing division leaders. (StatMuse)
This is a respectable record, but not elite given Toronto’s roster talent.
- Last ~11 games: The Leafs have had a strong stretch with a 7-2-2 record and a +11 goal differential (or slightly better in some samples). This means when they play well they win convincingly and control games offensively. (StatMuse)
- Expected Goals (last 7): ~24 goals for vs ~21 expected against in recent sample — again hinting at strong underlying offense. (StatMuse)
While exact up-to-date Corsi/xG% numbers aren’t publicly available in detail, proxy indicators (like NHL EDGE and NHL.com’s tracking) suggest:
- Toronto is middle-to-above average in shots and offensive danger at 5-on-5 based on general league metrics.
- Their offensive zone time is high, indicating puck possession and sustained pressure — both good signs of underlying control. (NHL)
Advanced analytic models like MoneyPuck typically show that teams with sustained possession and positive expected goals (xG) rates are more likely to outperform their record — or conversely, a worse record can signal issues outside of underlying play (goalie performance, luck, special teams, etc.). (General analytical principle)
- Goaltending has been spotty: Toronto’s top goalie numbers in the season stats include an x.920 SV% for one netminder, but other goalies around .884 in smaller samples — a wide gap. This indicates inconsistency in net that can swing outcomes. (Wikipedia)
- Historically, Toronto’s underlying talents (shots, expected goals, scoring chances) have often been better than their actual record when goaltending falters — similar to what analysts have described with Ottawa and other teams.
So
if Toronto’s goaltending is below league average on high-danger chances, then that
suppresses their record relative to their underlying play — exactly the phenomenon often highlighted with teams like Ottawa in analytics discussions.
Underlying Numbers vs Actual Results
Underlying play indicators suggest:
- Offense: Good control of possession and scoring chances; balanced offense with stars like Auston Matthews and William Nylander driving goals.
- Defense & Expected Goals Against: Likely decent, but goaltending variance and occasional defensive breakdowns can inflate the number of goals allowed relative to the chance quality.
- Goalie SV% swings: With one goalie trending above league average and others below, the actual goals allowed vs expected goals allowed gap likely exists.
In analytics terms:
- If Leafs’ expected goals for (xGF) > expected goals against (xGA) but actual goal differential is lower, that’s a classic indicator of underperforming record vs underlying performance. Based on minor proxies and recent positive goal differentials in stretches, that situation looks likely for Toronto this season.
- That means their actual record may be slightly worse than their underlying play would predict — possibly attributable to inconsistent goaltending or special teams bounces.
- Toronto sits mid-pack in standings despite having high-end talents and good scoring chance profiles — suggesting either some bad puck luck, goaltending regression, or clustering of goals allowed in certain games.
- Teams with similar underlying numbers but better goaltending/coaching execution tend to sit higher in the standings.
Underlying performance indicators for the Maple Leafs in 2025-26 suggest:
- Offense and puck possession are strong.
- Expected goals and chances generally favor Toronto.
- Actual record is solid but not elite given the roster.
- Inconsistent goaltending appears to be a limiting factor — preventing a better record from strong underlying play.
This is exactly the type of disconnect analysts refer to when saying a team’s underlying numbers look
better than their actual win-loss might imply — similar to how Ottawa’s underlying metrics were touted even when results lagged. (
StatMuse)
Is there anything else you want to ask it?
